From the Department of Molecular Medicine and Surgery Karolinska Institutet, Stockholm, Sweden
MECHANISM AND THERAPY OF
HEREDITARY ANGIOEDEMA TYPE III AND ROLE OF THE CONTACT SYSTEM IN
INFLAMMATORY DISEASES
Jenny Björkqvist
Stockholm 2014
All previously published papers were reproduced with permission from the publisher.
Published by Karolinska Institutet.
Printed by US-AB
© Jenny Björkqvist, 2014
ISBN 978-91-7549-706-8
Mechanism and Therapy of Hereditary Angioedema Type III and Role of the Contact System in Inflammatory Diseases
THESIS FOR DOCTORAL DEGREE (Ph.D.)
Friday 14 November 2014 at 09.00
By
Jenny Björkqvist
Principal Supervisor:
Professor Thomas Renné Karolinska Institutet
Department of Molecular Medicine and Surgery
Division of Clinical Chemistry Co-supervisor(s):
Docent Angela Silveira Karolinska Institutet Department of Medicine
Division of Atherosclerosis Research Professor Gunnar Nilsson
Karolinska Institutet Department of Medicine
Division of Clinical Immunology and Allergy Professor Lennart Lindbom
Karolinska Institutet
Department of Physiology and Pharmacology Division of Microvascular Physiology
Opponent:
Professor Marco Cicardi University of Milan
Department of Biological and Clinical Science Division of Internal Medicine
Examination Board:
Professor Carl-Fredrik Wahlgren Karolinska Institutet
Department of Dermatology Division of Medical Surgery 2 Professor Mona Ståhle Karolinska Institutet Department of Medicine Division of Rheumatology Professor Bo Nilsson University of Uppsala
Department of Immunology, Genetics and Pathology
Division of Clinical Immunology
ABSTRACT
Combinations of proinflammatory and procoagulant reactions are associated with a variety of disorders affecting the cardiovascular system. Vascular leakage contributes to the pathology of conditions such as, sepsis, allergy and anaphylactic reactions. Edema formation is the result of extravasated proteins and fluid and the peptide hormone bradykinin is considered to be one of the key mediators in the regulation of vascular leakage. Bradykinin is produced by the kallikrein-kinin system, which consists of factor XII, plasma prekallikrein, high molecular weight kininogen and C1 esterase inhibitor. Activated factor XII generates active prekallikrein (kallikrein), which cleaves kininogen, leading to the liberation of bradykinin. Activation of mast cells during allergic reactions mediates inflammatory responses, which cause increased vascular permeability. We report a new mechanism by which mast cell-released heparin increases vascular leakage. Upon allergen challenge mast cells release the negatively charged polysaccharide heparin that efficiently activates factor XII and initiates the kallikrein-kinin system. Heparin-driven kallikrein-kinin system activation culminates in bradykinin formation causing excessive vascular leakage in mice that are deficient in C1 esterase inhibitor, the major endogenous inhibitor of factor XII and kallikrein. These findings also have implications in anaphylactic and allergic diseases and we show that the factor XII–
driven kallikrein-kinin system critically contributes to the pathogenesis of anaphylaxis in both murine models and human subjects. The data indicate that heparin-initiated bradykinin formation plays a fundamental role in mast cell mediated diseases.
Hereditary angioedema (HAE) is a rare inherited disease that is characterized by acute swelling that involves the skin, extremities and mucosa. HAE types I and II are caused by deficiency in or dysfunctional C1 esterase inhibitor. In contrast, a third HAE variant exists in patients that have normal C1 esterase inhibitor (HAE III). HAE III is associated with a single point mutation at residue Thr309 in factor XII. However, the mechanism of HAE III was unknown. This study characterizes the mechanism and therapy of HAE III. HAE III patient- plasma and recombinant Thr309 mutated factor XII result in a double band or in a band with a lower molecular weight than wild-type factor XII in Western blotting. This is the consequence of a loss of glycosylation. The mutation in factor XII causes excessive activation of the kallikrein-kinin system resulting in enhanced production of bradykinin.
Addition of C1 esterase inhibitor dose-dependently blocked bradykinin production in HAE types I and II, but not in HAE III. We generated a fully humanized antibody (3F7) that specifically interferes with activated factor XII proteolytic activity. 3F7 inhibits activated factor XII-driven cleavage of high molecular weight kininogen in a dose dependent manner and interferes with aberrant kallikrein-kinin system-triggered bradykinin formation in HAE III plasma. We reconstituted factor XII deficient mice with recombinant human mutated factor XII and established an HAE III transgenic mouse that expresses human Thr309-mutated factor XII in the liver using Tet-off transgenic technology. Intravital confocal scanning microscopy and tracer extravasation-based methods show excessive bradykinin-mediated vascular leakage in both F12-/- mice reconstituted with mutated factor XII and in HAE III transgenic mice when challenged with factor XII-contact activator. Both a kallikrein inhibitor and 3F7 reduce edema in HAE III associated leakage in mice. This study characterizes the mechanism of HAE III and establishes factor XII inhibition as a novel therapeutic strategy to interfere with excessive vascular leakage in HAE III and potentially, other causes of edema.
LIST OF SCIENTIFIC PAPERS
The thesis is based on the following publications and manuscript:
I.
Björkqvist, J., Lecher, B., Maas, C., and Renné, T. Zinc-dependent contact system activation induces vascular leakage and hypotension in rodents.Biological Chemistry, 2013, 394, 1195-204.
II.
Oschatz, C., Maas, C., Lecher, B., Jansen, T., Björkqvist, J., Tradler, T., Sedlmeier, R., Burfeind, P., Cichon, S., Hammerschmidt, S., Müller-Esterl, W., Wuillemin, WA., Nilsson, G., and Renné, T. Mast cells increase vascular permeability by heparin-initiated bradykinin formation in vivo. Immunity, 2011, 34, 258-68.III.
Sala-Cunill, A., Björkqvist, J., Senter, R., Guilarte, M., Cardona, V., Labrador, M., Nickel, KF., Butler, L., Luengo, O., Kumar, P., Labberton, L., Long, A., Di Gennaro, A., Kenne, E., Jämsä, A., Krieger, T., Schlüter, H., Fuchs, T., Flohr, S., Hassiepen, U., Cumin, F., McCrea, K., Maas, C., Stavrou, E., and Renné, T. Plasma contact system activation drives anaphylaxis in severe mast cell–mediated allergic reactions. Journal of Allergy and Clinical Immunology, 2014, doi: 10.1016/j.jaci.2014.07.057IV.
Björkqvist, J., Oschatz,C., Lewandrowski, U., Schönig, K., Noethen, M.,Sickmann, A., Panousis, C., Maas, C., and Renné, T. Defective glycosylation of coagulation factor XII causes hereditary angioedema type III. Submitted Manuscript.
V.
Larsson, M., Rayzman, V., Nolte, MW., Nickel, KF., Björkqvist, J., Jämsä, A., Hardy, MP., Fries, M., Schmidbauer, S., Hedenqvist, P., Broomé, M., Pragst, I., Dickneite, G., Wilson, MJ., Nash, AD., Panousis, C., and Renné, T. A factor XIIa inhibitory antibody provides thromboprotection in extracorporeal circulation without increasing bleeding risk. Science Translational Medicine, 2014, 6, 222ra17Publications by the author, which are not included in the thesis
Björkqvist, J., Jämsä, A., and Renné, T. Plasma kallikrein: the bradykinin- producing enzyme. Thrombosis and Haemostasis, 2013, 110, 399-407 Björkqvist, J., Sala-Cunill, A., and Renné, T. Hereditary angioedema: a bradykinin-mediated swelling disorder. Thrombosis and Haemostasis, 2013, 109, 368-74
Björkqvist, J., Nickel, KF., Stavrou, E., and Renné, T. In vivo activation and functions of the protease factor XII. Thrombosis and Haemostasis, 2014, doi:
10.1160/TH14-04-0311.
Schürmann, D., Herzog, E., Raquet, E., Nolte, MW., May, F., Müller-Cohrs, J., Björkqvist, J., Dickneite, G., and Pragst, I. C1-esterase inhibitor treatment: preclinical safety aspects on the potential prothrombotic risk.
Thrombosis and Haemostasis, 2014, doi: 10.1160/TH13-06-0469.
Hansson, KM., Björkqvist, J., Deinum, J. Addition of prothrombin to plasma can result in a paradoxical increase in activated partial thromboplastin time.
Blood Coagulation and Fibrinolysis, 2014, doi:
10.1097/MBC.0000000000000161.
Hansson, KM., Björkqvist, J., Deinum, J. The Effect of Recombinant and Plasma Derived Prothrombin on Prothrombin Time (PT) in Human Plasma.
International Journal of Laboratory Hematology, 2014, doi:
10.1111/ijlh.12293.
CONTENTS
1 INTRODUCTION ... 1
1.1 The contact system ... 1
1.1.1 The intrinsic pathway of coagulation
... 2
1.1.2 The kallikrein-kinin system
... 2
1.1.3 Activation of the kallikrein-kinin system
... 3
1.2 The proteins of the kallikrein-kinin system ... 3
1.2.1 Factor XII (FXII)
... 3
1.2.2 Plasma prekallikrein (PPK)
... 4
1.2.3 C1 esterase inhibitor (C1INH)
... 6
1.2.4 High molecular weight kininogen (HK)
... 6
1.2.5 Bradykinin (BK)
... 6
1.3 Signaling pathways of BK ... 6
1.3.1 B1- and B2- receptors (B1R and B2R)
... 6
1.3.2 Signaling pathways
... 7
1.4 Degradation of BK ... 7
1.5 Mutation and deficiency in the kallikrein-kinin system ... 8
1.5.1 Mutations in FXII
... 8
1.5.2 Mutations in C1INH
... 9
1.6 Glycosylation ... 9
1.7 Allergy and anaphylaxis ... 10
1.7.1 Mast cells (MC) released heparin
... 10
1.8 Angioedema ... 11
1.8.1 Hereditary angioedema (HAE)
... 11
1.8.2 Hereditary angioedema type III (HAE III)
... 12
1.8.3 Other forms of angioedema
... 13
1.9 Inhibitors of the kallikrein-kinin system ... 14
1.9.1 FXII inhibitors
... 14
1.9.2 FXII inhibitory antibodies
... 15
1.9.3 PK inhibitors
... 15
1.9.4 PK inhibitory antibody
... 15
1.9.5 B2R antagonist
... 16
1.9.6 Recombinant and plasma derived C1INH
... 16
2 AIMS OF THE THESIS ... 17
3 EXPERIMENTAL PROCEDURES ... 19
3.1 In vitro methodology ... 19
3.1.1 Kallikrein-kinin activation assay (paper I, II, III, IV and V)
... 19
3.1.2 Chromogenic assay (paper I, IV and V)
... 19
3.1.3 Coagulation assay (paper II, III, IV and V)
... 19
3.1.4 Expression of Thr309Lys- and Thr309Arg-mutated and wild-type FXII (paper IV)
... 19
3.1.5 Inducible FXII_Thr309Lys expression in cells (paper IV)
... 20
3.1.6 Generation of HAE III transgenic mice using Tet-off system (paper IV)
... 20
3.2 In vivo methodology ... 21
3.2.1 Skin vascular leakage model (paper I, II and IV)
... 21
3.2.2 Miles edema model (paper II and IV)
... 22
3.2.3 FeCl3-induced arterial thrombosis model (paper IV)
... 22
4 RESULTS AND DISCUSSION ... 23
4.1 Establishment of methods for analyzing FXII-driven kallikrein-kinin activation (paper I)
... 23
4.2 Activation and the role of the plasma contact system in MC-mediated anaphylactic reactions (paper II and III)
... 25
4.3 Mechanism and therapy of HAE III (paper IV and V)
... 29
5 CONCLUDING REMARKS ... 39
6 POPULÄRVETENSKAPLIG SAMMANFATTNING ... 41
7 ACKNOWLEDGEMENTS ... 43
8 REFERENCES ... 45
LIST OF ABBREVIATIONS
A1 Apple domain 1
A2 Apple domain 2
A3 Apple domain 3
A4 Apple domain 4
ACE Angiotensin converting enzyme
ACEI-AAE Acquired angioedema related to angiotensin converting enzyme inhibitors
aPTT Activated partial thromboplastin time B1R B1 bradykinin receptor
B2R B2 bradykinin receptor
BK Bradykinin
bw Body weight
Ca2+ Calcium
[Ca2+]i Intracellular calcium concentration C1INH C1 esterase inhibitor
C1INH-AAE Acquired angioedema with C1 esterase inhibitor deficiency
CPB Cardiopulmonary bypass
CTI Corn trypsin inhibitor
DNP-HSA Dinitrophenyl-human serum albumin conjugate a-DNP-IgE IgE against dinitrophenyl
Dox Doxycycline
DXS Dextran sulfate
EA Ellagic acid
ECMO Extracorporeal membrane oxygenation EGF I Epidermal growth factor like domain I EGF II Epidermal growth factor like domain II FeCl3 Ferric chloride
Fib I Fibronectin domain type I Fib II Fibronectin domain type II
FITC-dextran Fluorescein isothiocyanate-dextran FXI Plasma coagulation factor XI FXII Plasma coagulation factor XII
FXIIa Activated plasma coagulation factor XII
a-FXIIa Activated plasma coagulation factor XII two-chain molecule b-FXIIa Activated plasma coagulation factor XII fragment
HAE Hereditary angioedema
HAE I Hereditary angioedema type I HAE II Hereditary angioedema type II HAE III Hereditary angioedema type III HK High molecular weight kininogen
IgE Immunoglobulin E
IH-AAE Idiopathic histaminergic acquired angioedema InH-AAE Idiopathic non-histaminergic acquired angioedema LC polyP Long-chain polyphosphates
MABP Mean arterial blood pressure
MC Mast cells
NO Nitric oxide
OSCS Oversulfated chondroitin sulfate PCK H-D-Pro-Phe-Argchloromethylketone
PK Plasma kallikrein
PPK Plasma prekallikrein
PS polyP Platelet-sized polyphosphates
rHA-infestin-4 Recombinant human albumin infestin-4 rtTA Reverse tetracycline controlled transactivator Serpin Serine protease inhibitor
Tet-on Tetracycline inducible expression system Tet-off Tetracycline controlled expression system tTA Tetracycline controlled transactivator
U Units
U-HAE Hereditary angioedema with normal C1 esterase inhibitor and of unknown origin
w/o Without
WT Wild-type
Zn2+ Zinc
1 INTRODUCTION
Under physiological conditions blood circulates in a closed system. Hemostasis is a balanced interaction of blood cell, vasculature and plasma proteins. A perfect hemostasis is when there is no bleeding and no thrombosis. Blood coagulation in parallel with inflammatory and repair provides a protective mechanism of the organism in response to vascular injury and pathological conditions. An example that accelerates the interactions of the reactants in these systems is a result of angioedema that could be driven by the kallikrein-kinin system, which is part of the contact system (Marder et al., 2013). This thesis is focused on the role of the contact system in inflammatory diseases (Figure 1).
1.1 THE CONTACT SYSTEM
The contact system is an enzymatic cascade in blood that exerts procoagulant and proinflammatory activities by the ability to activate the intrinsic coagulation pathway and the kallikrein-kinin system (Figure 1). The proteins assemble on cell surface like heparan- and chondroitin sulfate type proteoglycans (Colman et al., 1997; Renne et al., 2000; Renne and Muller-Esterl, 2001). The proteins of the contact system circulate in the bloodstream or are bound to the surface of different cell types, including endothelium, platelets, leukocytes and bacteria (Itakura et al., 2011; Renne et al., 2000; Renne et al., 2005b; Nickel and Renne, 2012). The contact system comprises five components: three serine proteases, factor XII (FXII), plasma prekallikrein (PPK) and factor XI (FXI), the non-enzymatic cofactor, high molecular weight kininogen (HK) and the C1 esterase inhibitor (C1INH), which is the major inhibitor of activated FXII (FXIIa) and activated PPK (plasma kallikrein, PK) (Colman and Schmaier, 1997).
Figure 1: The FXII-driven contact system. The contact system could activate the kallikrein-kinin system and the intrinsic coagulation pathway. Contact with negatively charged surfaces activates FXII on endothelial cells, leukocytes, bacteria and thrombocytes initiating procoagulant and proinflammatory proteolytic reactions. Activated FXII triggers activation of prekallikrein by FXIIa leading to generation of the vasoactive peptide BK by PK- mediated cleavage of HK, the kallikrein-kinin system. FXIIa generates fibrin formation through the FXI-mediated intrinsic coagulation pathway.
FXII is the primary initiator of the contact system cascade and binding to negatively charged surfaces results in a conformational change in the FXII zymogen. The conformational change results in small amounts of FXIIa (auto-activation). FXIIa converts PPK into its active form PK. This proteolytically active PK can feedback to activate additional FXII (hetero- activation), which amplifies the initial signal and also cleaves HK to liberate BK through the kallikrein-kinin system (Colman and Schmaier, 1997). BK is a nonapeptide hormone that belongs to the kinin family and initiates signaling cascades that lead to vasodilation, increased vascular permeability and tissue swelling. FXIIa also initiates other cascades such as the intrinsic coagulation pathway via its substrate FXI (Figure 1). Furthermore, in vitro FXIIa also has the capacity to drive the classic complement system pathway, and the fibrinolytic system (Bjorkqvist et al., 2013a; Marder et al., 2013).
1.1.1 The intrinsic pathway of coagulation
Blood coagulation in parallel with inflammatory and repair reactions are protective mechanisms of the organism in response to vascular injury. The coagulation system reacts quickly to stop blood loss from a damaged vessel wall. It involves the primary hemostasis (vasoconstriction, platelet adhesion and aggregation), secondary hemostasis (activation of coagulation factors and formation of fibrin) and fibrinolysis (activation of fibrinolysis proteins and lysis of the clot) (Furie and Furie, 1992). Impaired or excessive coagulation activity leads to an increased risk of hemorrhage (bleeding) or thrombosis (clotting). Macfarlane (Macfarlane, 1964) and Davie and Ratnoff (Davie and Ratnoff, 1964) were the first to describe the classical coagulation cascade or waterfall model. They consist of two converging enzymatic pathways initiated either by exposure of blood to a damaged vessel wall (extrinsic pathway) or by blood-borne components of the vascular system (intrinsic pathway). The intrinsic pathway of coagulation is initiated by FXII. FXIIa cleaves its substrate FXI to form active FXI, which in turn promotes coagulation via Ca2+-dependent activation of factor IX and with further activation resulting in conversion of prothrombin into thrombin.
Thrombin cleaves fibrinogen to form fibrin and activates multiple pathways in the vascular system (Furie and Furie, 1992).
To inhibit thrombosis anticoagulants are needed. One of the most common anticoagulants is heparin. In extracorporeal membrane oxygenation (ECMO) the blood comes into contact with prothrombotic artificial surfaces (such as the tubing and oxygenator) and anticoagulant treatment is crucial for preventing clots. ECMO is a form of cardiopulmonary bypass (CPB) used in intensive care. ECMO provides both cardiac and respiratory support to patients with severe lung or heart failure. Currently, unfractionated heparin is used as an anticoagulant (Sanchez et al., 1998; Sniecinski and Chandler, 2011) but despite intensive monitoring as well as surgical and pharmacological hemostatic therapies, life-threatening bleeding remains the major threat to ECMO patients (Conrad et al., 2005).
1.1.2 The kallikrein-kinin system
The kallikrein–kinin system is a part of the contact system that results in a proinflammatory response via BK-generation. BK production is started when FXII becomes auto-activated or hetero-activated. FXIIa cleaves surface associated plasma PPK to generate PK, which in
turn reciprocally activates further FXII molecules, thereby amplifying the initial signal (Bjorkqvist et al., 2013c) (Figure 1).
1.1.3 Activation of the kallikrein-kinin system
Activation of the kallikrein-kinin system and generation of BK is a consequence of FXII activators. FXII is activated by binding to negatively charged macromolecules, and these molecules can be separated into two groups: artificial substances such as silica-based materials (kaolin), negatively charged polymers (dextran sulfate, DXS), polyphenolic compounds (ellagic acid, EA) and endogenous substances including certain polymers, nucleotides, sulfatides, misfolded proteins, and some types of collagen or glycosaminoglycans (Cochrane and Griffin, 1982; Maas et al., 2008; Maas et al., 2011;
Muller et al., 2009; Oschatz et al., 2011; van der Meijden et al., 2009).
Recent studies have shown that misfolded protein aggregates only result in initiation of the kallikrein-kinin system without activating the intrinsic pathway of coagulation (Maas et al., 2008). Bacteria can also activate the kallikrein-kinin system when binding to the bacterial surface (Nickel and Renne, 2012). In addition, a molecule derived from platelets called polyphosphates (polyP) activates the contact system, resulting in the activation of both the kallikrein-kinin system and the intrinsic pathway of coagulation (Muller et al., 2009).
It has been shown that PPK could be triggered on endothelial cells independently of FXII activation (Motta et al., 1998). Intracellular chaperon heat shock protein 90 (Joseph et al., 2002), the mitochondrial prolylcarboxypeptidase (Shariat-Madar et al., 2002) have been described to activate PPK in cell culture systems, however their in vivo relevance remain a topic of future research. Extracellular carbonic anhydrase has shown to activate PPK in vivo (Gao et al., 2007) but more studies are necessary to elucidate the mechanism.
1.2 THE PROTEINS OF THE KALLIKREIN-KININ SYSTEM 1.2.1 Factor XII (FXII)
FXII is a serine protease and circulates in plasma as a single chain zymogen with molecular weight of 80 kDa. FXII has a concentration of 30-35 µg/ml (0.37 µM) and is primarily produced by hepatocytes with a half-life of 50 to 70 hours. FXII consists of 596 amino acids that code for the zymogen (Marder et al., 2013). The gene is composed of 13 introns and 14 exons and consists of an N-terminal fibronectin domain type II (Fib II, encoded by exon 3 and 4), an epidermal-growth-factor-like domain (EGF I, exon 5), a fibronectin domain type I (Fib I, exon 6), a second epidermal-growth-factor-like domain (EGF II, exon 7), a kringle domain (exon 8 and 9), a proline-rich region (55 amino acids, exon 9) and the catalytic domain (exon 10-14) (Cool and MacGillivray, 1987) (Figure 2). FXIIa consists of a disulfide- bond (Cys340-Cys367) linked heavy chain (52 kDa) and a light chain (28 kDa). The latter one mediates the enzymatic activity through the catalytic domain, which contains the catalytic triad Asp442, His393, and Ser554 (Colman and Schmaier, 1997).
FXII binds to negatively charged surfaces via its heavy chain. The surface-binding site of FXII is not precisely known, but several regions appear to contribute including: the distal N- terminal end (residue 1-28), the Fib II, the Fib I, the EGF II, the kringle domain and the proline-rich region (Marder et al., 2013).
Figure 2. Domain structure of FXII forms. The heavy chain, which contains the surface-binding region of FXII, consists of five domains Fib II, EGF I, Fib I, EGF II, kringle domain and the proline-rich region. The light chain harbors the enzymatic site of the serine protease. The two chains are connected by a single disulfide- bond (Cys340 and Cys367). Cleavage of the peptide bond Arg353-Val354 (arrow 1) results in two-chain activated FXII (α-FXIIa). The catalytic triad of FXIIa consists of His393, Asp442 and Ser544. Further proteolysis of the peptide bonds Arg343-Leu344 and Arg334-Asn335 (arrows 2 and 3) by PK, results in activated FXII-fragment (β- FXIIa) (Bjorkqvist et al., 2014).
FXII zymogen is activated by limited proteolysis cleaving the bond connection Arg353-Val354 and generating a two-chain molecule, FXIIa (α-FXIIa, Figure 2). FXII could be activated through two different principels. It is either activated when binding to negatively charged surfaces, which induce a conformational change (auto-activation) (Cochrane et al., 1973;
Samuel et al., 1992) or activated by other proteases such as PK (hetero-activation). Zinc (Zn2+) ions bind to FXIIa intermediate states and increase susceptibility for auto-activation by stabilizing conformations in the activation reaction (Bernardo et al., 1993). Further cleavages of the peptide bonds Arg334-Asn335 and Arg343-Leu344 of α-FXIIa by PK split the heavy from the light chain and result in activated FXII-fragment (β-FXIIa, Mw of 28-30 kDa, Figure 2).
Although β-FXIIa retains its proteolytic ability to activate PPK it is no longer able to bind to negatively charged surfaces. β-FXIIa has lost its capacity to convert FXI to active FXI and could not promote thrombosis (Maas et al., 2011; Marder et al., 2013).
1.2.2 Plasma prekallikrein (PPK)
PPK is a glycoprotein that is predominantly synthesized in the liver but has also been found in the epithelial cells of the kidneys, adrenal glands and placenta. The calculated molecular weight for the protein product of the mRNA is 79.5 kDa and the apparent molecular weight is approximately 85-88 kDa, suggesting that it likely reflects two different glycosylation forms.
PPK is secreted into the circulation as a single chain zymogen (Colman and Schmaier,
1997). PPK has a concentration of ~580 nM (50 µg/ml) in plasma and circulates mostly as a complex with HK (75-90%). PPK mRNA codes for a single chain polypeptide of 638 amino acid residues including an N-terminal signal peptide sequence of 19 residues. The mature PPK protein sequence of 619 residues is composed of five domains. The N-terminal portion consists of four tandem repeats of 84 to 85 residues each, called "apple" domains (designated A1, A2, A3 and A4). The four apple domains follow 248 residues that comprise the catalytic domain of the protein (amino acids 372 – 619, Figure 3). The PPK is stabilized by an arrangement of 18 disulfide bridges and is further modified by five N-linked complex- type oligosaccharides, two of which are located in the heavy chain portion and three on the light chain (Bjorkqvist et al., 2013a; Colman and Schmaier, 1997).
Figure 3: Domain structure of PPK. The primary translation product of 638 residues is cleaved at the NH2- terminus by signal peptidase, which detaches the leader peptide consisting of 19 amino acid residues. The N- terminal portion consists of four tandem repeats, called "apple" domains (designated A1 to A4) and the catalytic domain. Arrows emphasize the cleavage sites for activating the protease. Activated FXII cleaves PPK at a single site to generate a heavy chain and a catalytic light chain connected by a single disulfide bridge. The relative positions of the residues of the catalytic triad are His415, Asp464 and Ser559 (Bjorkqvist et al., 2013a).
PPK shares high homology with FXI and the proteases are 58% identical on the amino acid level, however they differ in their gross structures. PPK is a monomer whereas FXI is a homodimer. In contrast to PPK, FXI has a single free Cys residue in A4 (position 321) that forms a disulfide bond with the same residue in a second FXI polypeptide. Both PPK and FXI form tight 1:1 complexes with HK with KDs of 12 nM and 18 nM, respectively (Marder et al., 2013). In PPK, the apple domain A2 has been identified as the major binding segment for the HK domain 6, with domains A1 and A4 contributing to high- affinity binding (Herwald et al., 1993; Renne et al., 1999; Renne et al., 2002b). PPK is anchored to cell surfaces by binding to its substrate HK. PPK and PK bind to HK with similar affinity, indicating that proteolytic activation of PK does not interfere with the stability of the PPK/HK complex (Beaubien et al., 1991; Colman et al., 1985; Renne et al., 1999; Renne et al., 2002a).
The major physiological activator of PPK is FXIIa, which cleaves a single peptide bond (Arg371-Ile372) to generate a two-chain molecule with a heavy chain of 371 and a light chain of 248 residues held together by a disulfide bridge between Cys364 on A4 domain and Cys484 on the catalytic domain (Chung et al., 1986). Cleavage at Arg371-Ile372 results in a conformational rearrangement of the light chain and generates the catalytic active serine protease.
1.2.3 C1 esterase inhibitor (C1INH)
C1INH is a member of the family of serine proteinase inhibitors (serpins) and is a suicide inhibitor that inhibits proteases from the kallikrein-kinin system (FXIIa and PK) and also the complement proteases C1, the first complement of the complement system and MASP-1 and MASP-2 proteases in the lectin pathway. C1INH consists of 478 amino acids and contains an N-terminal domain (113 amino acids) and a serpin domain (365 amino acids).
The domains are connected via two disulfide bonds (Bos et al., 2002). C1INH has a molecular weight of 105 kDa and is highly glycosylated (Zuraw and Curd, 1986). Serpins contain a reactive peptide loop, which is important for interaction with its target proteases.
For the function of serpins, the most crucial amino acids in the loop are called P1–P1´
residues, and in C1INH it is position 444-445 (Bos et al., 2002). The protease cleaves the bond of P1–P1´ in the reactive center of the C1INH that induces a rearrangement of the protein and also forms a covalent binding between the P1 and the protease (such as FXIIa or PK) (Davis et al., 2008). The newly formed complex results in an inactive protease.
1.2.4 High molecular weight kininogen (HK)
HK is a 120 kDa α-globulin with a plasma concentration of 65-130 µg/ml. The heavy chain consists of domains 1 through 4 and domain 4 contains the BK region. The light chain consists of domain 5 and 6. The extreme C-terminal portion of domain 6 has the ability to form non-covalent complexes with either PPK/PK or FXI, and through domain 5 has the ability to interact with negatively charged surfaces with its histidine-rich region (Marder et al., 2013; Scott and Colman, 1980).
1.2.5 Bradykinin (BK)
Kininogens are the precursor of kinins, and one example of kinin is BK. BK is released by proteolysis of HK via PK and evokes inflammatory reactions. In the complex HK/PK, PK cleaves the HK peptide bond Arg371-Ser372 (Colman and Schmaier, 1997) and this step produces a two-chain HK form where BK is still attached to HK heavy chain (Mori et al., 1981; Nishikawa et al., 1992). A subsequent second PK-mediated cleavage at Lys362-Arg363 releases BK completely from HK. HK is then a two-chain protein that consists of a 65 kDa heavy chain and is linked via a disulfide bond to a 56 kDa light chain (Colman and Schmaier, 1997). The vasoactive proinflammatory BK has the amino acid sequence Arg1-Pro2-Pro3- Gly4-Phe5-Ser6-Pro7-Phe8-Arg9. Free BK is rapidly degraded and by cleaving Arg at position 9 of BK the active metabolite desArg9-BK is formed (Leeb-Lundberg et al., 2005).
1.3 SIGNALING PATHWAYS OF BK
1.3.1 B1- and B2- receptors (B1R and B2R)
BK acts through two distinct kinin receptors: B2-receptor (B2R) and B1-receptor (B1R). They are G protein-coupled receptors and are in most cells coupled to Gq and Gi (Leeb-Lundberg et al., 2005). These receptors mediate many pathophysiological functions of kinins, including edema formation, increased vascular permeability and capillary leakage, regulation of blood
pressure, hypotension, induction of fever, trans-endothelial cell migration and inflammation in different organs following injury (Leeb-Lundberg et al., 2005).
B1R is generally expressed at a low level and is largely up regulated by cytokines such as interleukin-1β (Kuhr et al., 2010; Leeb-Lundberg et al., 2005). Thus, B1R is exposed on cell surface in response to injury or inflammation. In contrast, B2R is constitutively expressed on various cell types, such as vascular smooth muscle cells, endothelial cells, and cardiomyocytes (Shukla et al., 2006). Kinins distinguish between these two receptors. BK primarily binds to B2R and des-Arg9-BK (degradation product of BK) primarily binds to B1R (Bhoola et al., 1992; Leeb-Lundberg et al., 2005).
1.3.2 Signaling pathways
Kinin-binding to B1R or B2R increases intracellular calcium [Ca2+]i, which generates arachidonic acid release resulting in prostacyclin and nitric oxide (NO) production. These products diffuse from the endothelium to the smooth muscle, followed by further activation of secondary mediators resulting in relaxation, decreased blood pressure and increase in blood flow. BK stimulates B2R (Gq) causing the activation of phospholipase C (PLC) leading to the formation of diacylglycerol (DAG) and inositol 1, 4, 5-triphosphate (IP3) which increases [Ca2+]i. [Ca2+]i can activate multiple signaling cascades, including the phospholipase A2 pathway where arachidonic acid is released leading to production of prostacyclin (Leeb- Lundberg et al., 2005). [Ca2+]i is also as a potent stimulator of endothelial nitric-oxide synthase (eNOS) and results in transient NO production. B1R interacts with Gq and Gi by desArg9-BK stimulation in which it acts through many of the same signaling pathways as B2R. Even though B1R and B2R are coupled to similar cellular signaling pathways, there is a difference in pattern. In vascular smooth muscle cells, stimulation of B2R leads to transient production of NO, whereas B1R stimulation leads to high and sustained NO production (Kuhr et al., 2010; Leeb-Lundberg et al., 2005).
Additional downstream signaling effects of BK include activation of protein kinase C (PKC).
PKC results in vasodilator-stimulated phosphoprotein (VASP)-mediated disassembly of cortical cytoskeletons (Benz et al., 2008) and of vascular endothelial (VE)-cadherin junctions in endothelial cells. Dependent on the cell type BK can induce proliferative and anti- proliferative responses and BK stimulation can also activate multiple transcription factors that are involved in tissue injury and inflammation (Leeb-Lundberg et al., 2005). NO and prostacyclin are thought to be major endothelium-derived vasodilators. However, all these factors have a role in BK-driven vasodilation and increased permeability (Shigematsu et al., 2002).
1.4 DEGRADATION OF BK
Non-receptor bound BK is rapidly degraded by multiple endo- and exopeptidases (kininases) (Skidgel, 1992) and has short half-life of <1 minute in plasma (Marder et al., 2013). One of the major kininases is kininase I (carboxypeptidase N) that removes the C-terminal Arg residue from BK. The resulting peptide des-Arg9-BK looses its affinity for binding to B2R however mediates its effects via activation of B1R. The digestion of des-Arg9-BK by kininase
II, also known as angiotensin converting-enzyme (ACE), results in release of the pentapeptide Arg-Pro-Pro-Gly-Phe and two dipeptides Ser6-Pro7 and Phe8-Arg9 (Bjorkqvist et al., 2013c; Sheikh and Kaplan, 1986). Other kinin-degrading enzymes are aminopeptidase P, dipeptidyl peptidase IV and neutral endopeptidase (Bjorkqvist et al., 2013c). Increased levels of BK in patients with various underlying diseases are difficult to monitor. There are, however, two conditions that are associated with high levels of BK namely hereditary angioedema (HAE, caused by a deficiency of C1INH) and severe infectious diseases (Cicardi et al., 2014; Frick et al., 2007).
1.5 MUTATION AND DEFICIENCY IN THE KALLIKREIN-KININ SYSTEM 1.5.1 Mutations in FXII
Defects in the FXII gene are known to cause loss of activity and/or deficiency of FXII.
Mutations either reduce FXII plasma levels or affect FXII secretion (Schloesser et al., 1997).
Some mutations reduce the enzymatic activity of FXIIa; these are located either in or close to the serine protease catalytic triad His393-Asp442–Ser544 (Schloesser et al., 1997). In 1955 the first patient, John Hageman, with FXII deficiency was described (Ratnoff and Margolius, 1955). Congenital deficiency in FXII (Hageman trait) is an autosomal recessive trait.
Deficiency in FXII, similarly to the kallikrein-kinin system proteins (PPK and HK) deficiency, prolongs one of the most common diagnostic tests aPTT (a coagulation assay that depends on contact system activation) but the individuals with contact factors deficiency do not get spontaneous or injury-related increased bleeding. On the other hand, deficiency in other coagulation factors such as: tissue factor (Bugge et al., 1996), factors VII (Rosen et al., 1997), VIII and IX (Hemophilia A, B) results often in spontaneous bleedings (Colman et al., 1975; Ratnoff and Margolius, 1955; Sollo and Saleem, 1985). FXII knock out mice are not able to form stable thrombi and have suppressed BK levels, this suggest that FXII is important for thrombus formation and inflammation (Iwaki and Castellino, 2006; Kleinschnitz et al., 2006; Renne et al., 2005a).
One of many mutations in FXII is FXII Bern mutation, which results in a secreted but dysfunctional protein and PK-cleaved FXII Bern is not able to activate FXI and/or PPK (Wuillemin et al., 1991). Another mutation is FXII Washington DC which has a Cys571-to- Ser571 substitution that leads to total loss of procoagulant activity in vitro (Miyata et al., 1989).
FXII Locarno is also secreted but is dysfunctional due to an Arg353 substitution. The mutation alters the FXIIa/PK recognition site in FXII and abolishes zymogen activation by limited proteolysis (Hovinga et al., 1994). Mutations that are located outside the catalytic domain and result in FXII deficiency are in domain Fib II and EGF (Kanaji et al., 2001).
Beside non-functional FXII, there are other mutations resulting in edema formation. One of these mutations is in position Thr309, which is located in the C-terminal proline-rich portion (residues 296-349) of the FXII heavy chain that mediates contact to other proteins and zymogen FXII surface binding (Citarella et al., 1996). This mutation is found in HAE III patients (described below). Moreover, a FXII gene deletion of 72 base pairs (starting at Lys305) was identified in two unrelated families with edemas (Bork et al., 2014). Additionally, another novel FXII gene mutation in the proline-rich region of FXII is a duplication of 18 base
pairs. This duplication causes the repeated presence of 6 amino acids (p.298–303) (Kiss et al., 2013). So far there are only mutations in the proline-rich region that corresponds to edema.
1.5.2 Mutations in C1INH
Mutations in C1INH result in non-functional or non-secreted protein. Upon activation of FXII and PPK, mutated C1INH is dysfunctional or insufficient to inhibit these proteases and the kallikrein-kinin system becomes activated. Mutation in C1INH is related to the disease HAE (see below). C1INH is also the major inhibitor of C1 complex in the complement system, which is part of the immune system. Mutated C1INH allows for activation of the initial phase of the complement system (classical- and lectin- pathway) and results in a reduction of plasma levels of complementary component C4 (Bork, 2014).
1.6 GLYCOSYLATION
Most of all secreted proteins are glycosylated and changes in the glycosylation pattern could affect the properties in the protein. Glycosylation is a modification when glycans attach to proteins or lipids and is a form of post-translational modification. The addition of carbohydrates has several functions such as folding and stabilizing of the protein. There are five different types of glycosylations, and N-linked glycosylation is the most common one followed by O-linked. The difference between the two is that the N-linked type has a glycan attachment to a nitrogen atom of an Asn side-chain that is present as a part of Asn-X- Ser/Thr, where X could be any amino acid except for proline, while the O-linked type has a glycan attachment to oxygen of Thr or Ser side-chain. O-linked glycosylations do not need a specific consensus sequence to occur. Nevertheless, previous studies have shown that O- linked glycosylation of Ser or Thr is more common if Ser/Thr residues are present in clusters, or are located in areas rich in proline or alanine residues (Preston et al., 2013).
Congenital disorders affecting glycosylation are rare but almost 50 different have been identified (Preston et al., 2013). Mutations leading to the loss or gain of a single specific glycosylation site have been implicated in diseases. A mutation in C1INH (C1INH Ta, (Rosen et al., 1971) has a deletion of nucleotides encoding Lys251. This amino acid deletion results in alteration of the sequence Asn-Lys-Ile to Asn-Ile-Thr thus creating a new potential N-linked glycosylation site and results in a non-functional C1INH (HAE II see below). The carbohydrate addition and/or the amino acid deletion result in altered conformation and function. Glycosylation leading to increased contact system activation is found for example in Lewis rats. They are more susceptible, than Buffalo and Fischer rats, to developing chronic intestinal and systemic inflammation by injection of group A streptococci (Sartor et al., 1996).
Furthermore, in Lewis rats a B2R antagonist (Icatibant) attenuated experimental inflammatory arthritis (Uknis et al., 2001). The difference between the rat strains is a single nucleotide substitution from Ser511 to Asn511 in HK in Lewis rats, presumably due to an extra N-glycosylation and lack of an O-glycosylation. Lewis rats have an increased rate of HK cleavage by PK (Isordia-Salas et al., 2003). This shows that glycosylation has an important function regarding both increased and loss of protein function. The FXII protein contains 16.8% carbohydrate, including 4.2% hexose, 4.7% hexosamine, and 7.9% N-
acetylneuraminic acid (Fujikawa and Davie, 1981). The heavy chain contains a proline-rich region (33% of the sequence) and includes six of the eight carbohydrates chain linkages of FXII, all of which are O-linked (Cys276-Arg334) including Thr309 (McMullen and Fujikawa, 1985). Thr309 is mutated in HAE III (Cichon et al., 2006).
1.7 ALLERGY AND ANAPHYLAXIS
Allergy is a hypersensitive disorder of the immune system and anaphylaxis is the most severe form of allergic reactions. It is a serious allergic or hypersensitive reaction, which has a rapid onset and can sometimes be fatal. Anaphylactic reactions represent an immunologic response to certain allergens resulting in a sudden, systemic degranulation of mast cells (MC) and basophils (Brown et al., 2013; Galli and Tsai, 2012; Kemp and Lockey, 2002) (Figure 4). Some of the most common trigger factors are: food, medications, insect venoms and other allergens. Anaphylaxis affects as much as 1-15% of the population (Wood et al., 2014). The reaction typically manifests with a broad range of symptoms such as hypotension, vascular leakage, bronchial constriction, cardiac arrhythmias, as well as gastrointestinal and skin manifestations.
Figure 4. Activation of several cascades caused by anaphylaxis. Anaphylaxis is commonly mediated through immune IgE-dependent mechanisms. Additionally non-immunologic or idiopathic mechanisms exist that cumulatively lead to activation of MC and basophils with mediator release for example heparin, modified from (Sala-Cunill et al., 2014).
1.7.1 Mast cells (MC) released heparin
MC are part of the immune system and are localized in the connective tissue throughout the body. They are granulocytes, a type of leukocytes, containing granules in their cytoplasm that could be released upon stimuli and cause inflammation. MC can be activated and degranulate through different mechanisms: via toll-like receptors, which are important for direct pathogen recognition, through complement products such as C3a and C5a and via IgE antibody interaction, through the receptor FcεRI. The allergen/IgE complex initiates intracellular signaling in the MC with release mediators that are vasoactive, particularly histamine, serotonin and proteoglycans (Dawicki and Marshall, 2007; Marshall, 2004;
Nilsson et al., 1996) (Figure 4). The mediators cause vasodilation, increase capillary leakage and can rapidly produce edema. Antihistamines (antagonists to histamine receptor) or corticosteroids are therapeutically used to treat allergic reactions and edema formation that are associated with aberrant MC activity (Theoharides and Kalogeromitros, 2006).
However, there are some patients that do not respond to these drugs.
MC secretory granules also contain highly sulfated proteoglycans, such as heparin. In vivo heparin is only synthesized in MC and contributes to the morphology and a storage capacity of their secretory granules. Heparin is a highly sulfated linear polysaccharide with repeating disaccharide units of 1-4-linked iduronic acid and glucosamine. Heparin is negatively charged and the average disaccharide unit has 2.7 sulfate groups (Capila and Linhardt, 2002).
1.8 ANGIOEDEMA
The definition of angioedema is edema localized in the subcutaneous and submucosal tissue, caused by local vasodilation and transient increase in vascular leakage due to release of vasoactive mediators. Urticaria is characterized by wheals, edema of the upper and mid-dermal skin layers, whereas angioedema is fluid leakage from deeper vessels (Bork, 2014). Urticaria and angioedema have many features in common; the edema is relatively short-lived in the skin and could involve other organs. Both belong to numerous disease entities and occur through various pathogenetic pathways. Quincke was the first who described angioedema as a separate entity in 1882 and named it angioneurotic edema that could still be referred to as Quincke edema (Quincke, 1882). Some years later Osler gave the first comprehensive description of angioedema where he called it “Hereditary angioneurotic edema” (Osler, 1888). The disease was later named HAE. Donaldson and Evens identified C1INH as the defective protein that is involved in the disease described by Osler (Donaldson and Evans, 1963). After that, angioedema research has been focused on revealing the patho-mechanism related to C1INH deficiency that eventually was shown to be a result of the release of BK (Cicardi et al., 2014).
Plasma levels of the peptide hormone BK are elevated during the swelling attacks in HAE patients. In acute episodes of HAE the mechanisms that result in increased vessel leakage is a result of excessive BK formation due to C1INH deficiency (Cugno et al., 2003; Joseph et al., 2008; Leeb-Lundberg et al., 2005; Schapira et al., 1983).
1.8.1 Hereditary angioedema (HAE)
Patients with HAE suffer from episodic swellings that can be fatal. The clinical symptoms include recurrent skin swelling, abdominal pain attacks, tongue swelling, and upper airway edema. The swelling typically lasts between 2–5 days (Davis, 2008; Zuraw, 2008). HAE is a rare inherited disease and has a prevalence of 1:10 000 to 1:100 000 and affects individuals from all races. Urticaria does not occur at any time in any of these patients (Cicardi and Agostoni, 1996; Longhurst and Cicardi, 2012). The increased vessel leakage is a result of excessive BK formation due to pathological activation of the FXII-driven plasma kallikrein- kinin system and plasma levels of BK are elevated during the swelling attacks (Cugno et al., 2003; Joseph et al., 2008; Schapira et al., 1983).
HAE is an autosomal dominant disorder and are present in different types. HAE I and II are affected by mutations in one of the two alleles in the SERPING1 gene coding for the C1INH
protein, and there are almost 300 mutations registered, http://hae.enzim.hu/. HAE I is caused by mutations leading to deficiency of C1INH (quantitative defect) and the mutations are randomly distributed throughout the SERPING1 gene. The mutations are point mutations, large rearrangements, including partial deletions and duplications (Bowen et al., 2001; Kalmar et al., 2003; Stoppa-Lyonnet et al., 1987). The mutations in type II result in a secreted C1INH but with a misfolded formation causing a reduced enzymatic activity (functional defect). Most mutations in HAE II are clustered in exon 8 of the SERPING1 gene, which encodes the active center (P1–P1´ residues) or hinge region of C1INH protein (Donaldson and Bissler, 1992). Despite having one normal allele, HAE I and II patients have lower than 50% of the physiological levels of C1INH. This is probably due to faster consumption of C1INH (Quastel et al., 1983). HAE has a high prevalence of de novo mutations, which account for around 25% of the cases (Pappalardo et al., 2000; Tosi, 1998).
1.8.2 Hereditary angioedema type III (HAE III)
HAE III does not differ clinically from HAE I or II, however, the patients have normal C1INH levels and functions (Bork et al., 2000). This type mostly affects women and was originally associated with estrogen intake. However a few male patients have been identified with HAE III questioning elevated estrogen levels to be causative for the disease (Charignon et al., 2014; Marcos et al., 2012; Riedl, 2013). Genome wide linkage analysis revealed a single point mutation in FXII. Two missense mutations at position Thr309 (1032C→A: Thr exchange to Lys, FXII_Thr309Lys, or 1032C→G: Thr exchange to Arg, FXII_Thr309Lys) have been identified (Dewald and Bork, 2006), which leads to increased FXII activity (Cichon et al., 2006). An in vitro study has shown that FXII_Thr309Lys mutation is a gain-of-function mutation, and that it has increased enzymatic activity however the FXII plasma levels are normal (Cichon et al., 2006). HAE III is also named FXII-HAE, however to be consistent with the original nomenclature (Bork et al., 2000; Cichon et al., 2006) the subtype of HAE with normal C1INH (Cicardi et al., 2014) that is associated with FXII mutations is here referred to as HAE III.
Bork and coworkers reported trigger factors that resulted in edema in 35 patients with HAE III. The most common ones included: acute trauma or physical pressure on the affected area, and after dental procedures. Some patients also reported emotional stress and ingestion of various spices and herbs as suspected triggers (Bork et al., 2009b; Riedl, 2013).
The swellings in HAE III patients are highly variable regarding triggering factors, severity, frequency, and localization but are mainly localized to the face or the extremities. Each episode lasts between 2-5 days. In some of the women the clinical symptoms were initiated by oral contraceptives, hormonal replacement therapy or pregnancy, but others were not affected by these conditions (Cicardi et al., 2014). HAE III patients should avoid triggers such as exogenous estrogen therapy and ACE inhibitors (described below) (Riedl, 2013).
Diagnosing HAE III is very challenging. HAE III is an autosomal dominant disease with low penetrance: asymptomatic carriers are >90% in male gender and around 40% in female (Bork et al., 2007; Marcos et al., 2012; Riedl, 2013; Vitrat-Hincky et al., 2010). Most HAE III patients originate from Europe mainly from Germany, Spain and France but patients have
also been reported from Canada, UK, Australia, Hungary and Morocco (Baeza et al., 2011;
Bork et al., 2014). The criteria for HAE III are listed in table 1.
Table 1: Recommended diagnostic criteria for HAE III reproduced from Zuraw et al. (Zuraw et al., 2012)
• A history of recurrent angioedema in the absence of concomitant hives or concomitant use of a medication known to cause angioedema
• Documented normal or near normal C4, C1INH antigen, and C1INH function
• Plus one of the following:
1. Demonstration of a FXII mutation associated with the disease
2. A positive family history of angioedema and documented evidence of lack of efficacy of chronic high-dose antihistamine therapy (cetirizine at 40 mg/day or the equivalent, for at least 1 month and an interval expected to be associated with three or more attacks of angioedema)
There are no controlled investigations of acute or preventative treatments making the medical management challenging. However, there has been significant progress in the development of therapeutic targets for HAE I and II over the last decade. The novel drugs are recombinant or plasma derived C1INH (Zuraw et al., 2010b), plasma kallikrein inhibition (Ecallantide, DX-88) (Cicardi et al., 2010b) and B2R antagonists (Icatibant) (Cicardi et al., 2010a). Replacement therapy with intravenous infusions of C1INH concentrate shortened the length of and frequency of swelling attacks in HAE I and II patients (Zuraw et al., 2010b).
In contrast to C1INH infusions, the drugs Icatibant and Ecallantide can be injected subcutaneously in an acute attack in HAE I and II patients, resulting in a relief the symptoms (Cicardi et al., 2010a; Cicardi et al., 2010b). In comparison to established therapies in HAE type I and II not much is known about treatment of type III but according to published case reports inhibition of kallikrein-kinin system and BK formation have shown to be beneficial.
Although published experience is limited, the B2R antagonist Icatibant has been reported to be efficient for the treatment of acute symptoms in several patients (Boccon-Gibod and Bouillet, 2012; Bouillet et al., 2009; Marcos et al., 2012). Ecallantide has also been reported successful in treating angioedema attacks in HAE III patients (Cronin and Maples, 2012).
Interestingly, despite normal measured C1INH quantity and function, the use of plasma- derived human C1INH during acute attacks has been reported effective in some cases, although reports of inefficacy also have been described. (Bork et al., 2009b; Bouillet et al., 2007; Marcos et al., 2012; Vitrat-Hincky et al., 2010). For long-term prophylaxis, progesterone, danazol and tranexamic acid have been used to treat HAE III with variable success however the steroid drugs have significant side effects (Bork et al., 2000; Herrmann et al., 2004; Saule et al., 2013; Vitrat-Hincky et al., 2010).
1.8.3 Other forms of angioedema
There are other forms of angioedema which are: Idiopathic histaminergic acquired angioedema (IH-AAE), Idiopathic non-histaminergic acquired angioedema (InH-AAE), Acquired angioedema related to angiotensin converting enzyme inhibitors (ACEI-AAE), Acquired angioedema with C1 inhibitor deficiency (C1-INH-AAE) and Hereditary angioedema with normal C1 inhibitor and of unknown origin (U-HAE) (Cicardi et al., 2014).
IH-AAE is an acquired disease with no identified cause. The patients present angioedema
that can be defined as histaminergic and these patients respond to high dose of antihistamines used prophylactically on a daily basis.
InH-AAE is a type of angioedema, which is a nonhereditary form and could not be treated with antihistamines. BK is probably involved in InH-AAE, however the experimental evidence is still limited (Cicardi et al., 2014).
ACE is one of the major degradation proteins of BK. ACE inhibitor (a drug against hypertension) results in elevated plasma levels of BK, which is the consequence in ACEI- AAE. Analysis of large cohort of patients administered with ACE inhibitor shows that angioedema occur in <0.5%. However, because of the large number of people taking these medications, ACE inhibitors are the leading causes of drug-induced angioedema. It has also been shown that genomic and plasma variability of the degradation proteins interferes with the BK catabolism and results in angioedema (Cicardi et al., 2014).
C1-INH-AAE is an acquired disease, which has no mutations in C1INH or any family history of angioedema. Studies from patient plasma indicate consumption of C1INH, some of the complement components (C1 and C4) and activation of the kallikrein-kinin system. C1-INH- AAE can be associated with C1INH deficiency caused by autoantibodies that neutralizing C1INH function. (Cicardi et al., 2014)
U-HAE patients have no mutations in C1INH or FXII genes and no other genetic defects that have be identified. U-HAE is inherited and during attacks they do not respond to corticosteroids or antihistamines (Cicardi et al., 2014).
1.9 INHIBITORS OF THE KALLIKREIN-KININ SYSTEM 1.9.1 FXII inhibitors
The recombinant FXIIa inhibitor rHA-infestin-4 is based on the fourth domain of the nonclassic Kazal-type serine protease inhibitor from the midgut of the insect Triatoma infestans fused to human albumin (Hagedorn et al., 2010). Intravenous infusion of the inhibitor prior to the challenge protects mice and rats from FeCl3-induced arterial thrombus formation, ischemic stroke (transient middle cerebral artery occlusion model) (Hagedorn et al., 2010), silent brain ischemia (Chen et al., 2012) and lethal pulmonary embolism (Muller et al., 2009). At high concentrations rHA-infestin-4 also inhibits plasmin and modestly factor Xa (Xu et al., 2014).
Another recombinant inhibitor Ixodes ricinus Contact Phase Inhibitor (Ir-CPI), a Kunitz-type protein from the salivary gland of Ixodes ricinus inhibits FXIIa, PK, and FXIa and provides protection from venous and arterial thrombus formation in mice (Decrem et al., 2009).
In murine models H-D-Pro-Phe-Arg-chloromethylketone (PCK) irreversibly inhibits the amidolytic activity of FXIIa and PK, and provided similar protection from ischemic stroke as in F12-/- without an increase in therapy-associated bleeding (Kleinschnitz et al., 2006).
Pretreatment with PCK markedly reduced cerebral infarction in the transient middle cerebral artery occlusion model (Kleinschnitz et al., 2006). Furthermore, PCK protected mice from platelet polyP-induced edema formation (Muller et al., 2009).
The specific FXIIa inhibitor Corn trypsin inhibitor (CTI) attenuated the prothrombotic properties of catheters in rabbits (Yau et al., 2012).
Other protein inhibitors, such as cabbage seed protease inhibitor (Carter et al., 1990), pumpkin seed inhibitor CMTI-V (Hojima et al., 1982) and Ecotin (Ulmer et al., 1995) block FXIIa in plasma, but also inhibit other proteases including thrombin, activated factor Xa, or XI, PK and plasmin.
1.9.2 FXII inhibitory antibodies
The monoclonal anti-FXII zymogen antibody 15H8 reduced fibrin formation and platelet accumulation in a collagen-coated vascular graft in baboons (Matafonov et al., 2014) and is directed to FXII zymogen. However, the antibody binding characteristics, its epitope and mechanisms of inhibition are not currently known.
Another antibody is 3F7, which is a recombinant fully human antibody that binds with high affinity (KD=6.2 ± 0.2 and 4.0 ± 0.1 nM for human and rabbit FXIIa, respectively) into the enzymatic pocket of FXIIa and specifically inhibits FXIIa activity and FXIIa-driven coagulation/inflammation in human, mouse and rabbit plasma (Larsson et al., 2014).
1.9.3 PK inhibitors
The drug Aprotinin (Trasylol) is an antifibrinolytic agent that was primarily used in complex and/or redo cardiac surgery as an addition to decrease postoperative bleeding and to reduce organ damage caused by hypotension. Aprotinin, a bovine pancreatic trypsin inhibitor, is a small protein, which inhibits several serine proteases including PK. However in 2008 the safety of Aprotinin was question because of a clinical trial showed a higher rate of death in patients receiving Aprotinin compared to other antifibrinolytic agents. The data in the clinical trail (Blood Conservation Using Antifibrinolytics in a Randomized Trial, BART) were not reproduced by independent trials and were only slightly above the statistical significance level but the drug was withdrawn from the market (Fergusson et al., 2008).
Ecallantide (DX-88, Kalibitor) is a novel recombinant PK inhibitor identified in a phage display-based screen containing the first Kunitz domain of human tissue factor pathway inhibitor variants. Ecallantide is a recombinant 60 amino acids protein that was selected on the basis of its affinity and specificity for PK (Williams and Baird, 2003). In a double-blind trail with HAE patients presenting with acute attacks Ecallantide was reported to give a significantly better outcome than placebo with regards to rapid relief and severity of symptoms. Ecallantide is injected subcutaneously and is now a drug for HAE patients used in acute attacks and is only approved by FDA (Cicardi et al., 2010b).
1.9.4 PK inhibitory antibody
DX-2930 was developed using phage display to select a potent and highly specific human antibody inhibitor of PK. Antibody therapeutics givens the potential for target specificity and they have a long serum half-life. Moreover, preclinical studies reveal DX-2930 to exhibit a
prolonged half-life in circulation (16-21 days), and that this translates into a prolonged capacity to inhibit the kallikrein-kinin system. Data support that DX-2930 can be used as a prophylactic inhibitor of PK in kallikrein-kinin system-mediated edema in vivo (Kenniston et al., 2014). DX-2930 has just completed a phase I study showing no toxicity symptoms in humans. The pharmacokinetic and pharmacodynamic data showed evidence for a long- acting biological effect and could be relevant to long-term prophylaxis for HAE patients (Chyung et al., 2014).
1.9.5 B2R antagonist
Icatibant (Firazyr) is a potent, specific and selective competitive B2R antagonist that is similar in structure to BK. It is short acting, and is composed of 10 amino acids, where five are synthetic and are resistant to fast degradation. In clinical studies administration of Icatibant was associated with symptomatic relief and is now used in acute attacks in HAE (Cicardi et al., 2010a; Cruden and Newby, 2008).
1.9.6 Recombinant and plasma derived C1INH
There are three different C1INH concentrates on the marked; recombinant C1INH (Rhucin), plasma derived C1INH (Berinert) and nano-filtered plasma derived C1INH (Cinryze) and some of these have been approved in Europe for decades but recently from the FDA.
Intravenous infusion of recombinant or plasma derived C1INH concentrate shortens the extent and duration of acute swelling attacks. Additionally when used for prophylaxis the concentrate reduced the frequency of acute attacks (Zuraw et al., 2010a).
2 AIMS OF THE THESIS
The overall aim of this thesis was to investigate the role of the contact system in inflammatory diseases with a focus on the mechanism and therapy of Hereditary angioedema type III in vitro and in vivo. More specifically, the studies can be divided into the following parts:
I) Establishment of methods for analyzing FXII-driven kallikrein-kinin activation
II) Activation and role of the plasma contact system in mast cell-mediated anaphylactic reactions
III) The mechanism and therapy of hereditary angioedema type III
3 EXPERIMENTAL PROCEDURES
3.1 IN VITRO METHODOLOGY
3.1.1 Kallikrein-kinin activation assay (paper I, II, III, IV and V)
Plasma was supplemented with increasing concentrations of the following triggers: dextran sulfate 500 000 Da (DXS, 1 pg/ml-100 µg/ml in 75 mM Tris pH 7.4, 10 µM ZnCl2), MC- derived heparins (1-1000 µg/ml) Ellagic acid (EA, 1.5 µg/ml), long- or short- chain polyphosphates (LC- SC- polyP, 10 µg/ml). Following incubation for 30 min at 37°C, the reaction was stopped by addition of Laemmli sample buffer, and 0.2-0.25 µl plasma was separated by polyacrylamide gel and analyzed by Western blotting using primary antibodies against FXII, PPK, HK, C1INH and FXI and horseradish peroxidase–coupled secondary antibodies. For some experiments FXII-deficient plasma was used.
3.1.2 Chromogenic assay (paper I, IV and V)
The enzymatic activity of FXIIa was photometrically measured during dextran, low molecular weight dextran sulfate 5 000 Da, DXS, EA, heparin, LC and PS polyP-induced activation using the chromogenic substrate S-2302 (D-Pro-Phe-Arg-p-nitroanilide, 1 mM) at an absorbance wavelength of 405 nm. The chromogenic substrate contains the peptide-p- nitroaniline and upon hydrolysis it is released and changes in color. The change in absorbance (ΔA/min) is directly proportional to the enzymatic activity.
3.1.3 Coagulation assay (paper II, III, IV and V)
The aPTT was measured on a Blood Coagulation System (BCS) or on a coagulometer by adding 50 µl of plasma samples, 50 µl Pathromtin SL, followed by incubation for 120 s at 37°C. Subsequently, 50 µl of a calcium chloride solution (25 mM) was added to start the reaction and the time was recorded.
3.1.4 Expression of Thr309Lys- and Thr309Arg-mutated and wild-type FXII (paper IV)
To generate FXII_Thr309Lys and FXII_Thr309Arg Site-directed mutagenesis was performed on human FXII cDNA (MIM ID: 610619) with the following primer:
5`-CCGAAGCCTCAGCCCAAGACCCGGACCCCGCCTCAG-3´ and 5`-CCGAAGCCTCAGCCCAGGACCCGGACCCCGCCTCAG-3´
resulting in the exchange of C at position 1032 to A (encoding for FXII_Thr309Lys) or G (encoding for FXII_Thr309Arg), in a pcDNA3 vector. Transient transfection of wild-type FXII and the mutants into HEK293 (human embryonic kidney) cells was done using Lipofectamine 2000 according to the manufacturer’s instructions. The supernatant was collected after 48 h and concentrated with Amicon Ultra centrifugal filters (30K).
